Hello all, I realize this has been somewhat discussed a couple of years ago but I believe the issue seems to be more complex and also, could be mitigated by now.
Given the fact that I’m portuguese and most of the works I’m mastering are portuguese as well, this issue presents itself quite regularly. Our alphabet uses characters like Ç, á, à, é, í, ã extensively. While the CD-text limitation are not really an issue, I’ve been battling with the metadata fields for a while and came to the conclusion that the following issue is most definitely an unresolved software bug.
When naming markers or tracks in the ALBUM panel I can write the special characters:
Now, if I use a third-party app to change the metadata, I can revert the title to its proper scheme but this makes the process very tedious. Specially if I need to re-render the files for some reason. But I noticed one interesting thing. If I save the audio montage and then try to rename the album track or maker name, when pressing backspace and deleting characters, once I reach a special character, it shows the unaccented character first and only then I can delete it. As if there is a hidden character encoding being deleted.
For example:
If I then re-render after performing this edit, then the metadata will show correct on windows:
But if I close the montage and reopen, it goes back to the wrong characters upon re-rendering.
Does anyone know if this has been reported / acknowledged by Steinberg?
What kind of metatada and how do you do to control it? In old metadata formats such as RIFF, there is no standard. Hence WaveLab does not use the operating system locale code, else the file can’t be read properly in other countries with a different locale code. For this, WaveLab uses unicode UTF-8.
But some software don’t follow that (good!) logic, hence won’t show that text properly. Those locale codes are ancient history and today unicode is the norm.
I understand what you’re saying but I don’t think this is an intentional feature. Wavelab not using operating system locale code isn’t something I can understand clearly, since it’s possible to use ISO 8859 and any of it’s parts.
In fact, if you look at the issue on the video I posted, you can see that after a re-edit of the marker name the title metadata gets exported correctly. Only after a re-open of the audio montage does it get back to the wrong characters and behavior.
As the behavior isn’t constant, I’m led to believe there is more than simple programming options here.
Those locale codes are ancient history and today unicode is the norm.
My understanding (maybe I’m wrong) is that unicode includes 8859 and all accented characters from Latin-1. Not sure why Wavelab should only use UTF-8 as metadata, specially for archival is extremely critical and titles should never be converted to the english language.
WaveLab has been doing this for more years than I can remember, and by conscious design. The reason for using Unicode is that it has long been the de facto standard.
Unicode is designed to represent almost all of the world’s writing systems, ensuring that your file’s metadata will be readable anywhere, as long as the reader supports Unicode—which is increasingly the case.
Unicode (UTF-8) is by far the best choice due to its compatibility, flexibility, and ability to meet current multilingual needs. Note that when writing metadata text, it’s not possible to specify, “This text is in ISO 8859, or Shift_JIS, or any other encoding.”
For instance, if you share a file with a German friend using ISO 8859 encoding, his German computer is likely to read the metadata incorrectly.
This is why Unicode is the only sensible choice.
Ok, it makes sense, so I must be confusing things here.
I’ve just tested re-opening an audio file in Wavelab after being renderes with wrong characters in the title and the metadata is corrrect, with the right accents in the RIFF tags. So can I assume this has to do with windows and the way it looks at the metadata encoding?
